It is very necessary for disease diagnosis, monitoring and treatment planning to locate and segment brain tumours from 3D MRI images accurately. 3D segmentation from MRIs means classifying each voxel in 3D space, it is very conducive to the relevant biological measurements and further analysis of the lesion. Until now, brain tumour segmentation from 3D biomedical images has been a challenging worldwide task due to the tumour features’ variousness, which varies part of U-Net and concatenates these features, which are upsampled to the same scale. To grasp the channel weight and ROIs, the bottleneck of the network is an improved dual path attention module, which convergence the advantages of channel attention and spatial attention. The proposed model has been validated in the online dataset of BraTS 2018. The mean dice score of enhancing tumours is 0.772. The mean dice score of the whole tumour is 0.907. The mean dice score of the tumour core is 0.819. The effectiveness of the proposed method is proved by quantitative and qualitative evaluation.

This paper considers the hybrid flow shop scheduling problem with two types of sequence-dependent setup times, one depending on the order of jobs and the other depending on both the order of jobs and the machines where the jobs are allocated. Three constructive heuristic algorithms based on the NEH algorithm are developed with the object of minimizing the total energy consumption cost. The first algorithm, called EPRA, obtains the order of jobs using the energy consumption cost based on the processing time. The second algorithm, called ESRA, obtains the order of jobs using the energy consumption cost based on the setup time. The third algorithm, called ESPRA, takes advantage of both EPRA and ESRA by constructing fictitious jobs with ESRA and acquiring the sequence of fictitious jobs with EPRA. Due to the drawback of NEH algorithm for the FFS problem proposed in this paper, all three heuristic algorithms use EST rules to allocate the jobs to the machines. Two lower bounds are developed to evaluate the heuristic algorithms proposed in this paper. A computational experiment based on the split-plot design is developed to carry out the factor analysis, algorithm analysis, lower bounds analysis, and CPU running time analysis. The analysis reveals that all main factor interactions except the algorithm are significant; the ESPRA algorithm is generally outstanding among the three algorithms; LB1 is better than LB2; the CPU running time of ESPRA algorithm is significantly reduced compared to those of EPRA and ESST algorithms.

As with the continuous improvement of the workshop automation rate and the importance in energy consumption, more and more enterprises not only need to make scheduling decision on production equipment, but also need to consider whether the scheduling of transportation equipment supports scheduling decisions on workshop production. At the same time, because both workshop production scheduling decision and transportation scheduling decision are NP-hard problems, it is necessary to design an efficient algorithm to improve productivity of the workshop. In order to solve this problem, firstly, based on the analysis of the problem structure, production environment and optimization objectives, a "manufacturing-transportation" multi-objective joint scheduling optimization mathematical model is established. By converting the energy consumption into the total transportation time objective of the transportation equipment, both total transportation time and makespan are taken as the optimization objectives. Secondly, based on the design idea of memetic algorithm (MA), non-dominated sorting genetic algorithm-Ⅱ(NSGA-II) is employed as the basis framework of our new developed algorithm. An effective discrete encoding scheme of MO-MA, a new initialization method for initial population and a neighborhood search mechanism based on critical path are incorporated into our new proposed algorithm. Then the parameter design of the algorithm is completed through variance analysis. Finally, the proposed algorithm is compared and analyzed with other algorithms in the dimension of hypervolume and Set Coverage (SC), and advantages of the algorithm in solving this problem are verified.